Research On Heterogeneous Coexisting Networks Of Aloha And CSMA:Modeling,Optimization And Application

The explosion of mobile data in 5G networks has led to a large gap in demand for spectrum.To alleviate it,the full use of unlicensed spectrum becomes even more necessary.As a common resource,all kinds of wireless access technologies could operate on the unlicensed spectrum,so the coexistence of heterogeneous networks of different standards has become an inevitable trend.Aloha and CSMA are two representative random access schemes.Due to the coupling between them,it is difficult to accurately analyze the performance of heterogeneous coexisting networks by theoretical modeling.Although existing research has performed performance analysis on the Aloha and CSMA coexisting network,key issues such as throughput performance limit,fair coexistence and performance optimization theory of this heterogeneous coexisting network have still not been resolved.In order to solve these issues,the thesis launched the following research.Firstly,according to the characteristics of Aloha and CSMA coexisting networks and the coupling relationship of cross-network parameters,a theoretical model of heterogeneous random access network is proposed in the thesis.Based on the model,explicit expressions of the throughput performance and the probability of successful transmission of heterogeneous networks are derived,which clearly show the impact of network parameters on throughput performance.Compared with the existing research models,the theoretical model has lower computational complexity and is more scalable.The analytical results show that the impact of coexisting counterpart on one's own network throughput is unbalanced,and the network throughput of CSMA is greatly affected by Aloha network.Considering that the throughput of Aloha and CSMA coexisting networks are degraded by the dynamic changes of network parameters,this thesis further analyzes the throughput optimization based on the proposed theoretical model.According to whether each network only maximizes its own throughput or jointly maximizes the total throughput of coexisting networks under the fairness constraint,this thesis proposes corresponding non-cooperation and cooperation optimization schemes,and both schemes provide corresponding adaptive adjustment strategies of optimal transmission parameters.Compared with the non-cooperative optimization scheme,the cooperative optimization scheme takes into account the fairness of coexisting networks,and could optimize the total throughput performance of coexisting networks given the inter-network throughput ratio.Especially given the throughput fairness constraint,the total throughput performance could be improved by up to 29.7% under cooperative optimization scheme.Finally,this thesis applies the theoretical model and optimization scheme to the coexisting network scenario of unlicensed 5G new radio and WiFi,which solves the problems of lack of fairness and low total throughput performance caused by the existing coexisting scheme and standard parameter settings.Based on the existing coexisting scheme,an improved coexisting scheme based on non-Listening-before-Talk is proposed.According to the corresponding relationship between the theoretical model and the actual unlicensed 5G new radio and WiFi coexisting network scenario,the optimal parameter settings of the theoretical model are applied to the network scenario.Analytical and simulation results show that the optimization scheme of the model could not only guarantee the throughput fairness,but also keep the maximum total throughput of the coexisting network at a stable level,without significantly decreasing with network parameters.In summary,this thesis takes heterogeneous coexisting networks of Aloha and CSMA as the research object,and on this basis,a universal and scalable theoretical model is obtained by reasonable modeling analysis.In the heterogeneous coexisting networks,the corresponding optimization scheme is proposed to solve the problems of throughput degradation and fairness unbalance.It is applied to the coexisting network scenario of unlicensed 5G new radio and WiFi,which proves the superiority and universality of the theoretical model and optimization theory,and also provides theoretical guidance for the subsequent design of heterogeneous random access networks.